Is There a Null Hypothesis?

The null hypothesis significance testing that we're accustomed to doing is always aiming to sort of reject out, right? We could imagine flipping that test and thinking instead about what I'm really kind of clumsily calling, rejecting in. Imagine the null hypothesis defines this region that's outside where we would say, oh yeah, that's a real correlation. And then the alternative to that is that our correlation in the population actually falls somewhere between negative 0.10 and positive 0.10. Well if we flip the null and alternative hypotheses like that, we can actually use hypothesis testing procedures to try to test whether or not we could consider that correlation to be practically equivalent to zero.